The first round is with HR, who asks about your background and a bunch of short technical questions about concurrency, databases, microservices, etc. These are generally very simple questions.

The next round is live coding, followed by a technical interview and finally a design interview. There is usually 1-2 weeks between each stage, so the process is long.

Be prepared to be interviewed by a Russian individual with poor English, lacking people skills, and absolutely no interest in the interview. During my technical interview, the individual was clearly tired and annoyed, with a crying baby in the background. Throughout the entire interview, they did not look at the camera, instead working on another screen.

They do not even ask about your background. You will be given a task to do, and if you are lucky, they may pay attention to what you are doing or saying. There was very little interaction or interest to find out if you know about the subject. There is an element of luck involved in answering questions because if they do not hear what they want, they will not try to dig deeper or clarify with you.

If you do not pass any round, you will receive a generic reply from Revolut with no feedback. In my case, HR even ignored a follow-up email where I thanked them for the opportunity and kindly asked for feedback.

Finally, they do not hire for a specific role. So, even if you pass all four rounds, there is no guarantee you will be offered a job. You will only be placed into a "pool" and they will start looking for a suitable role for you within the company.

(This kind of makes sense after reading the reviews – they have a high attrition rate.)

In retrospect, it was not worth it and was likely the worst interview process I have experienced. The process took a long time with totally unclear selection criteria and no feedback from Revolut.

Simple load balancer live coding.

Simple money transfer function to complete.

How would you implement a balance transfer from one account to another in BD?

How would you improve BigQuery performance in a relational DB?

What is the CQRS pattern - pros/cons?

What stability patterns do you know?

What deployment strategies do you know?